基本的分治思想算法运用

在算法设计和解决问题中，分治策略中，递归地求解一个问题，在每层递归中应用如下3个步骤：

分解：将原问题分解为若干子问题，子问题与原问题形式一致，只不过规模更小

解决：递归地求解子问题，如果子问题够小，则直接求解，停止递归

合并：将子问题的解合并成原问题的解

实例：求一个数组的最大子数组的区间与和

A = [13,-3,-25,20,-3,-16,-23,18,20,-7,12,-5,-22,15,-4,7]

注：数组中需要有负数，否则无意义。实例场景：在一组已知的数据中求最大的数据差值与区间。

def MAXCROSSINGARRAY(A,low,mid,high):
    leftsum = -65536
    tempsum = 0
    i=mid
    while(i>=low):
        tempsum = tempsum+ A[i]
        if(tempsum>leftsum):
            leftsum = tempsum
            maxleft = i
        i=i-1
    rightsum = -65536
    tempsum = 0
    i=mid+1
    while(i<=high):
        tempsum = tempsum +A[i]
        if(tempsum>rightsum):
            rightsum = tempsum
            maxright = i
        i=i+1
    return maxleft,maxright,leftsum+rightsum

def MAXMINSUBARRAY(A,low,high):
    if(high==low):
        return (low,high,A[low])
    else:
        mid = (low+high)/2
        leftlow,lefthigh,leftsum = MAXMINSUBARRAY(A,low,mid)
        rightlow,righthigh,rightsum = MAXMINSUBARRAY(A,mid+1,high)
        crosslow,crosshigh,crosssum = MAXCROSSINGARRAY(A,low,mid,high)
        if(leftsum>=rightsum and leftsum >=crosssum):
            return leftlow,lefthigh,leftsum
        if(rightsum>=leftsum and rightsum >= crosssum):
            return rightlow,righthigh,rightsum
        else:
            return crosslow,crosshigh,crosssum

A = [13,-3,-25,20,-3,-16,-23,18,20,-7,12,-5,-22,15,-4,7]
low,mid,A_sum = MAXMINSUBARRAY(A,0,15)
print low,mid,A_sum

一般情况下使用暴力求解，运行时间为Ω（n2）,但是通过递归情况，最后当n!=1时，T（n）= 2T(n/2)+O(n)，n=1时，T（n）=O(1)